create-principles-skill
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill accepts a user-provided
[area]argument that is interpolated into a brief for thefigure-outskill. This creates a surface where untrusted input could potentially influence downstream agent actions during the investigation process. - Ingestion points: Input parameter
[area]defined in the YAML frontmatter and used in the instruction brief withinSKILL.md. - Boundary markers: Absent; the brief lacks specific delimiters or instructions to the model to ignore any embedded commands within the area name.
- Capability inventory: The skill has the capability to invoke the
figure-outskill, which performs broader project investigations and data processing. - Sanitization: No input validation or escaping mechanisms are present to sanitize the
[area]argument before it is passed to the secondary skill.
Audit Metadata